A photogrammetry-based methodology to obtain accurate digital ground-truth of leafless fruit trees

In recent decades, a considerable number of sensors have been developed to obtain 3D point clouds that great potential in optimizing management agriculture through the application precision techniques. order use data provided by these sensors, it is essential know their measurement error. this paper, methodology presented for obtaining cloud central axis training system defoliated fruit tree (Malus domestica Bork.) obtained from stereophotogrammetry techniques based on structure-from-motion (SfM) and multi-view stereo-photogrammetry (MVS). The was made set 288 photographs scene including ground truth which used generate digital model. resulting validated proven faithfully represent reality. bias model −0.15 mm 0.05 mm, diameters lengths, respectively. addition, allows small changes actual be detected as consequence wood dehydration process. Having an ground-truth basis validating other sensing systems vegetation characterization can make more informed decisions.